Research map / Learning theory and algorithms / Combinatorial and search optimization
Modular Robot Design Synthesis with Deep Reinforcement Learning
Julian Whitman, Raunaq Bhirangi, Matthew Travers, Howie Choset
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace IK-based heuristic best-first search with DQN-guided best-first search | modular design search · Modular manipulator design, 11 modules, 50 target sets | avg runtime to first feasible (min): 3.04 -> 0.26 | Improved |
| replace genetic algorithm with DQN-guided search | modular design search · Modular manipulator design, 11 modules, 50 target sets | avg complete robot evaluations to first: 138.12 -> 10.31 | Improved |
| replace genetic algorithm with DQN-guided search | modular design search · Modular manipulator design, 11 modules, 50 target sets | avg best cost after five minutes: 0.53 -> 0.52 | Mixed |
| replace genetic algorithm with DQN-guided search under torque constraints | modular design search · Modular manipulator design, 16 modules with torque constraint, 50 target sets | trials with no feasible robot: 28/50 -> 1/50 | Improved |
| replace best-first search with DQN search | modular design search · Modular manipulator design, 11 modules, 50 target sets | trials with no feasible robot: 16/50 -> 1/50 | Improved |
| condition the heuristic on obstacle locations (ablation) | DQN heuristic · modular design search | solution quality and runtime | Improved |
Closest papers by meaning
- Multi-Task Reinforcement Learning with Soft Modularization (NeurIPS 2020 · similarity 0.86)
- DittoGym: Learning to Control Soft Shape-Shifting Robots (ICLR 2024 · similarity 0.86)
- Deep Reinforcement Learning for Robotics: A Survey of Real-World Successes (AAAI 2025 · similarity 0.86)
- MeMo: Meaningful, Modular Controllers via Noise Injection (NeurIPS 2024 · similarity 0.84)
- Deep Generative Model for Mechanical System Configuration Design (AAAI 2025 · similarity 0.83)
- MorphVAE: Advancing Morphological Design of Voxel-Based Soft Robots with Variational Autoencoders (AAAI 2024 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
